Defect size is the first question to answer

The size of your cartilage defect is usually the first thing your surgeon will use to narrow your options. Not your age, not your sport, not your preference — size, measured in square centimetres, is the lead clinical variable because it determines whether your own knee can physically supply enough repair material.

Osteochondral autograft transfer (OATS or mosaicplasty) harvests cylindrical plugs of cartilage and bone from low-load areas of the same knee. Those donor zones are finite. For defects roughly below 4 cm², the technique is practical and the donor site can recover without meaningful lasting harm. Above that threshold, harvesting enough plugs begins to create cumulative damage in an otherwise healthy part of the joint — effectively converting one injury into two. This is not a matter of surgical preference; it is a biological supply limit.

The 2–4 cm² band is a genuine grey zone. Within it, both autograft and allograft can be considered, and the surgeon's direct assessment of available donor cartilage carries real weight. Below roughly 2 cm², autograft is generally the default. Above approximately 4 cm², a cadaveric (allograft) graft becomes the standard approach, because it imposes no volume restriction from the patient's own anatomy.

Before a first specialist appointment, it is worth asking whether defect size has been measured — ideally on MRI — because that single figure shapes the entire conversation that follows.

How autograft transfer (OATS) works and when it fits

During an OATS procedure, the surgeon uses a circular coring instrument to lift one or more cylindrical plugs — each containing both cartilage and the bone beneath it — from areas of the knee that bear little load during normal movement, typically the margins of the femoral condyle. Those plugs are pressed directly into the defect site, restoring the native cartilage surface and the structural bone layer in a single operation. Because harvesting and implantation happen together, there is no waiting period between stages, no cell culture, and no second procedure.

Mosaicplasty — placing several smaller plugs in a tile-like pattern — extends the treatable area somewhat, but the donor supply remains the binding constraint. Once the low-load zones are exhausted, any further harvesting starts to affect cartilage that does real mechanical work; the risks to the harvest site then outweigh the benefit to the repair site. That ceiling falls at roughly 4 cm², though the exact point depends on each patient's anatomy and what the surgeon finds at the time.

Within the appropriate size range, published long-term follow-up data suggest durable outcomes in approximately 78% of patients. That figure holds best for younger patients without a prior history of knee surgery; older age and previous operations on the same knee are independent risk factors for reoperation. When defect size and patient profile are well matched, autograft transfer is not a compromise — it is the procedure for which the strongest size-matched evidence exists.

How osteochondral allograft (OCA) works and when it becomes the standard choice

When a defect exceeds what a patient's own knee can supply, a donor graft from a cadaveric source removes that ceiling entirely. Osteochondral allograft transplantation (OCA) replaces the damaged cartilage and the underlying bone in a single operation using a size-matched graft from a tissue bank — there is no harvesting from the patient's knee, so the volume of the repair is limited only by the defect itself, not by anatomy.

Formal indications for OCA, as set out in a 2022 systematic review, are: symptomatic full-thickness cartilage lesions larger than 3 cm², defects that extend into the subchondral bone beneath the cartilage, and revision cases where a prior procedure has already failed. These are precisely the scenarios in which autograft transfer is either insufficient or no longer viable.

Survivorship data provide a useful benchmark. Approximately 89% of grafts remain intact at five years and around 75% at twelve years, with a mean failure rate of roughly 25% at that point and a reoperation rate of approximately 36% over the same period. When OCA does fail, most conversions are to total knee replacement (around 68%) or unicompartmental replacement — an outcome worth understanding before surgery, even if it remains a minority scenario.

The choice between dowel and shell surgical techniques is governed by lesion geometry and is a decision the operating surgeon makes, not the patient. Risks associated with cadaveric tissue — including immune response and disease transmission — are historically rare with modern tissue banking and processing, though recent case reports have documented occasional MRI-visible reactions in a small number of patients. This is worth raising in the preoperative consultation rather than a reason to avoid the procedure.

Patient and lesion factors that shift the decision

Defect size sets the frame, but several other variables shift the balance — and a well-prepared patient can raise each of these directly with their surgeon.

Lesion surface. Is the damage confined to one joint surface, or does it affect the opposing surface as well? Unipolar lesions — damage on a single surface — are consistently associated with better outcomes from OCA than bipolar involvement, where both sides of the joint are affected.

Patient age. Age under 30 is linked to better OCA survivorship. For autograft OATS, older age carries its own risk: it is an independent predictor of reoperation and longer-term failure.

Cause of injury. A traumatic origin — a sports collision, a fall, a twisting accident — carries a better prognosis than cartilage loss arising from gradual degeneration, for both procedures.

Time from symptoms to treatment. For OCA specifically, treatment within 12 months of symptom onset is associated with meaningfully better outcomes. Prolonged delay matters.

Lesion location. Where in the knee the defect sits is a clinically important consideration, particularly for OCA. Patellofemoral grafts — those involving the kneecap joint — carry a reoperation rate of around 83%, substantially higher than grafts placed on the femoral condyle or tibial plateau. This is not a reason to rule out surgery in the patellofemoral joint, but it is a difference patients deserve to understand before consenting.

What the preoperative workup involves

Neither graft procedure exists in isolation. Before surgery goes ahead — for autograft or allograft alike — the surgeon will assess the whole mechanical environment of the knee, not just the cartilage defect itself. This is not administrative caution; it is what makes outcomes reliable.

Three areas are non-negotiable in that assessment:

• Lower-limb alignment. If the leg is malaligned, load is distributed unevenly across the joint. A graft placed into a malaligned knee bears disproportionate stress and is more likely to fail early. Where malalignment is found, an osteotomy — a bone realignment procedure — is often carried out at the same time as grafting, or in a planned prior stage.
• Meniscal integrity. The menisci act as shock absorbers and help distribute force across the joint surface. A deficient or absent meniscus alters that load pattern in ways that directly affect graft survival. Meniscal status is assessed before any grafting decision is finalised.
• Ligamentous stability. An unstable knee — for instance, one with an unaddressed ACL or PCL injury — places abnormal shear forces on any repair. Instability may need correcting alongside or before grafting.

Where correctable co-pathologies are identified, the surgeon will map out whether to address them in the same operative episode or in a staged plan. Understanding this upfront helps set realistic expectations about what the overall treatment journey may involve.

Comparing outcomes honestly and finding the right specialist

Both procedures deliver broadly comparable medium-term success when matched correctly to lesion size and patient profile — published series place autograft OATS and osteochondral allograft in the 78–89% range, with no meaningful hierarchy between them when each is used within its correct indication. The clinical question, then, is less which procedure than which patient, which lesion, which surgeon.

Direct head-to-head trial data for the 2–4 cm² transition zone are limited. A 2023 systematic review (Trofa et al., cited extensively in the orthopaedic literature) confirmed comparable medium-term outcomes for both techniques, but noted that differences in patient and lesion selection between study arms prevent precise direct comparison. No large randomised controlled trial has enrolled patients specifically within this grey zone — which is why individual surgeon assessment carries weight that published algorithms alone cannot replace.

For anyone approaching a specialist consultation after reading this, the questions worth raising are: what is the measured defect area on MRI; which surface or surfaces are affected; is there co-pathology requiring concurrent correction; what would the failure pathway involve; and what is the realistic timeline to full loading? Understanding these variables means a patient can evaluate the answers they receive, not merely hear them.

Cartilage restoration consultants offering OCA and OATS practise across the UK. Search MSK is a specialist directory covering practitioners in this field — patients can filter by procedure area and region to find a consultant suited to their situation.